Strategies for Analyzing Online Networks

Live blogging workshop on online network analysis at UW, Digital Media Working Group (DMWG).

Participants: Kirsten Foot, The Coproduction of Online Structures, Associate Professor, UW; Rachel Gibson, The Study of Weblinks as Communicative Devices, Chair, New Media Studies, University of Leicester, UK; Clifford Tatum, Hyper-Lurk Analysis, PhD student, Communication, UW; and Mike Xenos, Making Sense of Link-Based Network Patterns, Associate Professor, Communication Arts, University of Wisconsin-Madison.
Kirsten:
Interested in a form of co-production that is not just about information but actually building structures produced by different types of actors — between organizations — some are created by hyperlinks (a structure for action). Following 9-11, government sites outlinked to Red Cross, community organizations, etc — this was a “first” — using the web to build structures for info and forms of action.

Research on fair trade movement/advocacy … the problem is how hard it is (systematically, methodologically) to find these actionable links. Find the traces of links and systematically characterize these links — and how best to represent that.

Action links to other sites [keg -> look for this behavior in mainstream press]

Can we annotate the automated link diagram with qualitative information from researchers.

— What artifacts do you identify as action?
Kirsten: we have a typology for things that can be called action.

— What would be a link that was not mobilizing?
Kirsten: most are informational.

— Is there any way of capturing traffic?
Kirsten: yes, by looking at log files … but not realistic for webstorm type of event.
Find in-bound links to action site as a proxy.

Clifford:
Seattle Urban Blogosphere
Interest is in collaborative knowledge production – not necessarily formal sense – how lay knowledge may be challenging formal knowledge. You can look at search engines – emergent — the more explicit way is wikipedia, which is intentional structure of knowledge. Blogs are something in-between — some intentionally as well as loose + self-organizing. Struggle is IDing consumers and their role in the knowledge production … not unlike lurkers on e-mail list.

Sept 2006 – 32 residential or mixed use bldgs in permit process/under construction/recently completed. Lots of urban development and planning. How to find a condo in the pre-sale stage? Discovered urban blogosphere after the fact. Highly distributed knowledge creation with “radical transparency” and it challenges “authoritative knowledge.”

Theoretically : “middle” media (Bennett, Bimber, Perretti – many-to-many) … destablization of information economy (Benkler)

Hard to understand the role of the reader/lurker.

Issue crawler analysis — reflects who is dominant in this space. Couple of large nodes are real estate agents.

GeoMapping – readership for Clifford’s blog is from around the world

SLOG – Vulcan Trolling on 2200 Westlake Blog. ‘Caught me’

Google Analytics can move people back to Blogger because of analytical tools. Check search engine referrals? [99% are from other blogs] Direct traffic is the 3rd biggest … coming from a bookmark or typing it in. A lot of innovators are sharing results, information.

Rachel:
Strategies of analysis on nanotechnology debate. Operationalising the communicative functions of hyperlinks. http://www.voson.anu.edu.au

Political communication thru hyperlinks has been studied thru two basic perspectives:
– site/actor-centric approach – focus is on organization
– contextual/structural approach – focus is relational questions (communication as power)

hub scores – authority scores

five benefits of hyperlinks (ackland and gibson, 2006)
– network building
– audience sharing
– info provision
– identity building/branding
– message amplification

How to measure?
– quantity of outbound links (function 1 and 2)
– quality/composition of outbound links (function 3 and 4)
– extent of link-back between set of sites (function 5)

[keg: -> it would be interesting to know structure of the link – an image? text? if text, what words?]
[keg: -> look at candidate websites and do a qualitative and quantitative analysis of outbound links — compare with party websites? ah! expand on ACM analysis and examine hosts/software/etc of candidates by party]

A reminder that all links are not necessarily endorsements – example of far left linking to far right site re differing narrative of mutual event

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s